Do Large Language Models Think Like the Brain? Sentence-Level Evidences from Layer-Wise Embeddings and fMRI

TL;DR

This study investigates the alignment between large language models and human brain activity during sentence comprehension, revealing that higher semantic layers in LLMs increasingly resemble neural responses, thus advancing understanding of their cognitive parallels.

Contribution

It systematically compares layer-wise embeddings of 14 LLMs with fMRI data, demonstrating how model performance improvements lead to brain-like hierarchical representations.

Findings

Higher semantic layers in LLMs show stronger correlation with brain activity.

Model improvements enhance the alignment with neural hierarchies.

LLMs exhibit brain-like patterns that emerge from deeper architectural features.

Abstract

Understanding whether large language models (LLMs) and the human brain converge on similar computational principles remains a fundamental and important question in cognitive neuroscience and AI. Do the brain-like patterns observed in LLMs emerge simply from scaling, or do they reflect deeper alignment with the architecture of human language processing? This study focuses on the sentence-level neural mechanisms of language models, systematically investigating how layer-wise representations in LLMs align with the dynamic neural responses during human sentence comprehension. By comparing hierarchical embeddings from 14 publicly available LLMs with fMRI data collected from participants, who were exposed to a naturalistic narrative story, we constructed sentence-level neural prediction models to identify the model layers most significantly correlated with brain region activations. Results show that improvements in model performance drive the evolution of representational architectures toward brain-like hierarchies, particularly achieving stronger functional and anatomical correspondence at higher semantic abstraction levels. These findings advance our understanding of the computational parallels between LLMs and the human brain, highlighting the potential of LLMs as models for human language processing.